Current unmet medical needs in the hemophilia field are mainly: (1) treatment of hemophilia patients with inhibitors (˜30% of hemophilia patients); and (2) long acting and efficacious coagulant factors (FVIII/FIX) and/or their replacement (bypass drugs) (WFH report 2012, Paris). The most widely used bypass drug for treating hemophilia patients with inhibitors is rFVII, which has major drawbacks such as risk of thrombogenicity, short half-life in plasma and high production cost. Antibodies against anti-coagulant factors, such as Tissue Factor Protein Inhibitor (TFPI), APC (Activated Protein C) and Antithrombin (AT) represent a new treatment paradigm. These antibodies not only bypass or reduce the need for FVIII or FIX coagulation factors in hemophilia patients with inhibitors, but also exhibit longer plasma half-life (which reduces the dosing frequency) and, thus, increases patient compliance. To date, there have been several antibody-based procoagulant drugs at the preclinical development or research stage, such as anti-TFPI and anti-APC.
AT is a major anticoagulant in human plasma. It inhibits thrombin, FXa and other serine proteases functioning in the coagulation pathway. It consists of 432 amino acids, is produced by the liver hepatocyte and has a long plasma half-life of three days (Collen, Schetz et al. 1977). The amino acid sequence of AT is well-conserved and the homology among cow, sheep, rabbit, mouse and human is 84%-89% (Olson and Bjork 1994). Although the primary physiological targets of AT are thrombin and FXa, AT also inhibits FIXa, FXIa, FXIIa, as well as FVIIa to a lesser extent. AT exerts its inhibition together with heparin. In presence of heparin the inhibition rate of thrombin and FXa by AT increases by 3 to 4 orders of magnitude from 7-11×103 M−1 s−1 to 1.5-4×107 M−1 s−1 and from 2.5×10−3 M−1 s−1 to 1.25-2.5 M−1 s−1 respectively (Olson, Swanson et al. 2004).
Unlike TFPI and APC which inhibit coagulation solely at the initiating stage and the amplification stage respectively, AT exerts its inhibition on coagulation at both the initiation and amplification stage. Therefore, blocking AT could have more potent pro-coagulant effect than blocking either TFPI or APC alone. Decreased AT levels and activity have been shown to correlate with increased thrombosis in human. Patients with AT deficiency tend to show recurrent venous thrombosis and pulmonary embolisms (van Boven and Lane 1997). Furthermore, homozygous AT knockout mice die in the embryonic stage with an extreme hypercoagulable state (Ishiguro, Kojima et al. 2000). A recent study shows that heterozygous AT knockout hema mice in which AT is reduced by 50% significantly have less blood loss and enhanced thrombin generation in a tail-clip bleeding model (Bolliger, Szlam et al. 2010).
AT is a glycoprotein with two isoforms based on differential glycosylation on Asn135, ATα and ATβ (Bjork 1997). ATβ lacks glycosylation at Asn135 and is a minor glyco-isoform representing 10% of human plasma AT. Asn135 is located adjacent to the initial heparin attachment site and constitutes part of extended heparin binding site after allosteric activation and D helix extension (dela Cruz, Jairajpuri et al. 2006). The lack of bulky-sized glycan at Asn135 affects ATβ activation profoundly in two ways: 1) a faster allosteric activation upon heparin binding required for inhibition of FXa and FIXa; and 2) extra accessible binding sites for higher affinity heparin binding for inhibition of FXa and thrombin by a bridging mechanism. Indeed, under physiological salt concentration, plasma-derived ATβ binds to heparin with a KD of 36+/−3 nm while ATα binds to heparin with a KD of 500+/−50 nm (Turk I V. et al., 1993). The higher affinity of ATβ for heparin leads to its preferential distribution to the sub-endothelial layer which is enriched in the heparin-like structure—glycosaminoglycan. Consequently, ATβ is proposed to play a major and potent role in inhibition of FXa and thrombin at the vascular injury sites (Carlson and Atencio 1982; McCoy A J, Pei X Y. et al. 2003; Turk B, Brieditis I. et al. 1997; Witmer M R, Hatton M W. 1991; Frebelius S, et al. 1996). The importance and stronger potency of ATO relative to that of ATα is also reported in clinical studies. In patients, the severity of AT homozygous mutations defective in heparin-binding is ameliorated by the beta form of AT (Martinez-Martinez, Navarro-Fernandez et al. 2012). In another study, a borderline level (˜70/of normal AT antigen and activity) of AT is compensated by the 20%-30% ATβ in plasma (Bayston, Tripodi et al. 1999).